feat: add Agones readiness supervisor core

2026-09-10 16:04:04 +00:00 · 2026-08-31 20:40:51 +01:00
parent 698413cd91
commit a0195987bd
3 changed files with 242 additions and 2 deletions
@@ -1209,8 +1209,8 @@ the local/CI/community transport, not a silent production fallback.
 | # | Task | Acceptance |
 |---|---|---|
 | 8.26 `[D:8.1,8.6,8.12]` | Portable Helm/Kustomize Fleets per build/EU/NA region; isolate provider edge/network/DNS/secret and SDR POP/cert/public-UDP overlays | Two provider fixtures render; labels select region/build/protocol/transport; each fixture documents Valve approval and externally reachable UDP mapping |
-| 8.27 `[D:8.26]` | Godot Agones REST adapter plus allocation-metadata watch and Go PID-1 supervisor scaffold; both bypass cloud behavior without SDK env; local SDK support | Native/existing Compose/CI remain functional; emulator exercises supervisor port discovery plus Get/Watch, Ready, Health, annotation and Shutdown |
-| 8.28 `[D:8.6,8.27]` | **Process-ready stage:** supervisor obtains dynamic port, launches Godot; static config/listen/Health succeed, then explicit Agones Ready—no roster/backend-registration prerequisite and no stdout probe | A detached unallocated process reaches Ready; a broken listener/config never does; Health reclaims a hung process |
+| 8.27 `[D:8.26]` | **IN PROGRESS.** Go supervisor package provides local-safe Agones REST discovery, dynamic `SDR_LISTEN_PORT`/`SDR_IP` injection, explicit process-ready probing and Ready transition; direct mode bypasses Agones | `server/supervisor/` covers allocated/direct startup and dynamic endpoint/Ready ordering; Godot Agones adapter, metadata watch, Health/annotation/Shutdown and emulator integration remain |
+| 8.28 `[D:8.6,8.27]` | **IN PROGRESS.** Supervisor separates explicit process-ready from Agones Ready and never scrapes stdout; allocated mode refuses to mark Ready without a configured readiness probe | `server/supervisor/` tests prove Ready follows the probe and direct mode remains functional; Godot readiness endpoint, detached-container and Health-reclaim integration remain |
 | 8.29 `[D:8.26,8.27]` | Separate ENet and Hosted-SDR dynamic/passthrough mappings; supervisor exports local `SDR_LISTEN_PORT` and external `SDR_IP`; validate POP/cert/firewall/NAT | Two isolated matches share a node; Agones-reported public endpoint receives relay traffic on the bound socket; ENet fixture remains independent |
 | 8.30 `[D:8.18,8.26,8.28,8.29]` | Atomic `GameServerAllocation` from Ready filtered by region/build/protocol/transport, attaching signed roster/non-secret config with bounded race retry | Duplicate commands yield one Allocated server; exhaustion or retry leaves no orphan; no client assignment is exposed merely because process is Ready |
 | 8.31 `[D:8.9,8.30]` | **Assignment-ready stage:** watch Allocated metadata, verify manifest/bindings, register hosted address, acknowledge backend; only then mint/expose client tickets | Modified/wrong manifest never reaches assignment-ready; clients cannot connect early; secrets never appear in metadata/args/logs |
@@ -0,0 +1,168 @@
+// Package supervisor contains the small PID-1 lifecycle boundary around an
+// allocated Godot process. The Agones client is HTTP-only so local/Compose
+// execution remains independent of the cloud SDK.
+package supervisor
+
+import (
+	"context"
+	"encoding/json"
+	"fmt"
+	"net/http"
+	"os"
+	"os/exec"
+	"strconv"
+	"strings"
+	"time"
+)
+
+type GameServer struct {
+	Status struct {
+		Address string `json:"address"`
+		Ports   []struct {
+			Name string `json:"name"`
+			Port int    `json:"port"`
+		} `json:"ports"`
+	} `json:"status"`
+}
+
+type Config struct {
+	Command      []string
+	Environment  []string
+	SDKBaseURL   string
+	ReadyURL     string
+	ReadyTimeout time.Duration
+	PollInterval time.Duration
+	HTTPClient   *http.Client
+}
+
+type Supervisor struct {
+	config Config
+	client *http.Client
+	cmd    *exec.Cmd
+}
+
+func New(config Config) (*Supervisor, error) {
+	if len(config.Command) == 0 || config.Command[0] == "" {
+		return nil, fmt.Errorf("supervisor command is required")
+	}
+	if config.ReadyTimeout <= 0 {
+		config.ReadyTimeout = 30 * time.Second
+	}
+	if config.PollInterval <= 0 {
+		config.PollInterval = 100 * time.Millisecond
+	}
+	if config.HTTPClient == nil {
+		config.HTTPClient = http.DefaultClient
+	}
+	return &Supervisor{config: config, client: config.HTTPClient}, nil
+}
+
+// Start launches the process and marks Agones Ready only after the explicit
+// readiness probe succeeds. No stdout/log scraping is used. With no SDK URL,
+// this is direct/Compose mode and the command is simply started.
+func (s *Supervisor) Start(ctx context.Context) error {
+	env := append([]string(nil), os.Environ()...)
+	env = append(env, s.config.Environment...)
+	if s.config.SDKBaseURL != "" {
+		port, address, err := s.assignedEndpoint(ctx)
+		if err != nil {
+			return err
+		}
+		env = append(env, "SDR_LISTEN_PORT="+strconv.Itoa(port), "SDR_IP="+address+":"+strconv.Itoa(port))
+	}
+	s.cmd = exec.CommandContext(ctx, s.config.Command[0], s.config.Command[1:]...)
+	s.cmd.Env = env
+	if err := s.cmd.Start(); err != nil {
+		return err
+	}
+	if s.config.SDKBaseURL == "" {
+		return nil
+	}
+	if err := s.waitReady(ctx); err != nil {
+		_ = s.cmd.Process.Kill()
+		return err
+	}
+	return s.sdkPost(ctx, "/ready")
+}
+
+func (s *Supervisor) Wait() error {
+	if s.cmd == nil {
+		return fmt.Errorf("supervisor has not started")
+	}
+	return s.cmd.Wait()
+}
+
+func (s *Supervisor) assignedEndpoint(ctx context.Context) (int, string, error) {
+	var server GameServer
+	if err := s.sdkGet(ctx, "/gameserver", &server); err != nil {
+		return 0, "", err
+	}
+	if len(server.Status.Ports) == 0 || server.Status.Address == "" {
+		return 0, "", fmt.Errorf("Agones returned no assigned endpoint")
+	}
+	for _, port := range server.Status.Ports {
+		if port.Port > 0 && (port.Name == "game" || len(server.Status.Ports) == 1) {
+			return port.Port, server.Status.Address, nil
+		}
+	}
+	return 0, "", fmt.Errorf("Agones returned no usable game port")
+}
+
+func (s *Supervisor) waitReady(ctx context.Context) error {
+	if s.config.ReadyURL == "" {
+		return fmt.Errorf("allocated mode requires an explicit readiness URL")
+	}
+	deadline := time.NewTimer(s.config.ReadyTimeout)
+	defer deadline.Stop()
+	for {
+		request, err := http.NewRequestWithContext(ctx, http.MethodGet, s.config.ReadyURL, nil)
+		if err == nil {
+			response, requestErr := s.client.Do(request)
+			if requestErr == nil {
+				_ = response.Body.Close()
+				if response.StatusCode >= 200 && response.StatusCode < 300 {
+					return nil
+				}
+			}
+		}
+		select {
+		case <-ctx.Done():
+			return ctx.Err()
+		case <-deadline.C:
+			return fmt.Errorf("process-ready probe timed out")
+		case <-time.After(s.config.PollInterval):
+		}
+	}
+}
+
+func (s *Supervisor) sdkGet(ctx context.Context, path string, target any) error {
+	request, err := http.NewRequestWithContext(ctx, http.MethodGet, strings.TrimRight(s.config.SDKBaseURL, "/")+path, nil)
+	if err != nil {
+		return err
+	}
+	response, err := s.client.Do(request)
+	if err != nil {
+		return err
+	}
+	defer response.Body.Close()
+	if response.StatusCode/100 != 2 {
+		return fmt.Errorf("Agones GET %s returned %s", path, response.Status)
+	}
+	return json.NewDecoder(response.Body).Decode(target)
+}
+
+func (s *Supervisor) sdkPost(ctx context.Context, path string) error {
+	request, err := http.NewRequestWithContext(ctx, http.MethodPost, strings.TrimRight(s.config.SDKBaseURL, "/")+path, nil)
+	if err != nil {
+		return err
+	}
+	response, err := s.client.Do(request)
+	if err != nil {
+		return err
+	}
+	defer response.Body.Close()
+	if response.StatusCode/100 != 2 {
+		return fmt.Errorf("Agones POST %s returned %s", path, response.Status)
+	}
+	return nil
+}
@@ -0,0 +1,72 @@
+package supervisor
+
+import (
+	"context"
+	"net/http"
+	"net/http/httptest"
+	"os"
+	"path/filepath"
+	"strings"
+	"testing"
+	"time"
+)
+
+func TestAllocatedStartInjectsDynamicEndpointAndCallsReadyAfterProbe(t *testing.T) {
+	ready := false
+	readyCalled := false
+	server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		switch r.URL.Path {
+		case "/gameserver":
+			_, _ = w.Write([]byte(`{"status":{"address":"203.0.113.9","ports":[{"name":"game","port":31001}]}}`))
+		case "/ready-probe":
+			if ready {
+				w.WriteHeader(http.StatusOK)
+			} else {
+				w.WriteHeader(http.StatusServiceUnavailable)
+			}
+		case "/ready":
+			readyCalled = true
+			w.WriteHeader(http.StatusOK)
+		default:
+			w.WriteHeader(http.StatusNotFound)
+		}
+	}))
+	defer server.Close()
+
+	ready = true
+	path := filepath.Join(t.TempDir(), "env.txt")
+	command := []string{"/bin/sh", "-c", "env > " + path}
+	s, err := New(Config{Command: command, SDKBaseURL: server.URL, ReadyURL: server.URL + "/ready-probe", ReadyTimeout: time.Second, PollInterval: time.Millisecond})
+	if err != nil {
+		t.Fatal(err)
+	}
+	if err := s.Start(context.Background()); err != nil {
+		t.Fatal(err)
+	}
+	if err := s.Wait(); err != nil {
+		t.Fatal(err)
+	}
+	contents, err := os.ReadFile(path)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if !strings.Contains(string(contents), "SDR_LISTEN_PORT=31001") || !strings.Contains(string(contents), "SDR_IP=203.0.113.9:31001") {
+		t.Fatalf("dynamic endpoint not injected: %s", contents)
+	}
+	if !readyCalled {
+		t.Fatal("Agones Ready was called before process-ready probe")
+	}
+}
+
+func TestDirectModeDoesNotRequireAgonesReadiness(t *testing.T) {
+	s, err := New(Config{Command: []string{"/bin/sh", "-c", "exit 0"}})
+	if err != nil {
+		t.Fatal(err)
+	}
+	if err := s.Start(context.Background()); err != nil {
+		t.Fatal(err)
+	}
+	if err := s.Wait(); err != nil {
+		t.Fatal(err)
+	}
+}